AIMS: We propose the use of a multi-center, prospective, cohort design with standardized, open, outcome variable evaluation technique to non-experimentally study patients aged 18-75 years with chronic renal insufficiency (CRI) defined as125I-Iothalamate GFR 30-70cc/min/1.73m2. The primary goals of the Chronic Renal Insufficiency Cohort Study will be: (1) to determine the risk factors for accelerated decline in renal function;and (2) to evaluate the incidence and risk factors for cardiovascular disease (CVD). METHODS: We propose to recruit 500 patients of which at least 30% are African American and 50% have diabetic nephropathy. We will recruit the study participants from 12,729 prescreened patients with CRI from our integrated clinical research units consisting of: The University of Michigan Health System (n=5,807);Veteran Affairs Medical Center, Ann Arbor (n=358);St. Johns Health System, Detroit (n=l,430);and The Wellness Plan, Detroit (n=5,134). Ongoing clinicalresearch efforts of the proposing team (includingCRI Pilot Study of 105 patients and AASK) have documented an overall participationpotential of 17.6% in the 12,729 prescreened subjects which enables a final pool of 2,321 subjects. Our recruitment strategies and retention/adherence techniques have been previously tested and logistically refined. Proposed primary outcomes variables are: (1) the mean rate of GFR decline;(2) time to end stage renal disease (ESRD);(3) composite fatal and nonfatal CVD events;and (4) all-cause mortality rate. Key explanatory variables to be studied include: Left Ventricular imaging (Echocardiography);Carotid Intima/Media Thickness (B-mode ultrasonography);and Cardiac Dysautonomia (["C]-hydroxyephedrine {HED} Positron Emission Tomography). All key study measurements will be obtained during a single 24-hour in-patient GCRC yearly visit supplemented with standardized,regularly scheduled, interval ascertainment non-visit procedures. ANALYSIS PLAN: For analytic efficiency, we propose a stratified recruitment scheme that allows broad distributionover the entire inclusionGFR range. We propose a two-stage analysis of the slope of GFR consisting of within-patientGFR slope estimates and across-patient slope analysis to evaluate the determinants of variationin the slope using repeated measures analysis and linear regression analysis, respectively. Time to event analyses will be carried out until ESRD, de novo/recurrent CVD event or death using a multivariateCox regression technique. A sample of 3,000 patients will be sufficient to yield 90% power to detect differences in rates of decline of GFR as small as 12% to 15% of the mean, and to detect increases as small as 16% to 40% in the rates of events (ESRD, death, CVD). SIGNIFICANCE: This study will lead improved patient care and generate new hypotheses about potential predictors of CRI progression and CVD complications. The most promising of these hypotheses will require testing in randomized clinical trials.